Cross-layer adaptation of spectrally efficient Raptor codes for video multicasting over mobile broadband networks

This paper explores the use of rateless application layer Forward Error Correction (AL-FEC) based on Raptor codes in a spectrally efficient way for real-time video multicasting over mobile broadband networks. The link between channel errors at the PHY, MAC and application layers and the required Raptor code parameters for quasi-error free video transmission is investigated. Low Raptor code rates are shown to place high demands on the limited bandwidth of a wireless channel. The selection of Modulation and Coding Scheme (MCS) is extended to take into account the channel resources required to accommodate the additional Raptor overheads. A cross-layer optimization approach and a novel metric are proposed to select the Raptor code parameters and the MCS mode jointly to maximize wireless transmission efficiency. Simulation results show that 166% more bandwidth is required by a non-Raptor enabled multicast WiMAX system, assuming a Mobile Station (MS) speed of 10km/h, in comparison to the optimized Raptor AL-FEC multicast WiMAX system. Results show that more bandwidth is required at lower MS speeds. Furthermore, the optimized Raptor AL-FEC system is shown to eliminate packet loss, to enhance the link budget by up to 6dB, which quadruples multicast coverage in LoS locations, and to increase the total goodput by 67%, in comparison to a non-Raptor enabled multicast WiMAX system.